Sulfur Toleration in Gasoline - ACS Publications - American Chemical

Sulfur Toleration in Gasoline. Gustav Egloff and C. D. Lowry, Jr. Universal. Oil Products Company, Chicago, III. THERE is constant pres- sure on the o...
1 downloads 0 Views 694KB Size
August, 1928

I,VD VSTRIAL ALVDEXGINEERING CHEMISTRY

839

Sulfur Toleration in Gasoline’ G u s t a v Egloff a n d C. D. Lowry, Jr. UNIVERSAL OIL PRODLTTS COMPANY, CHICAGO, ILL.

T h i s s t u d y presents an analysis of gasoline c o n s u m p HERE is constant presgeologist that 62 per cent of t i o n in relation to t e m p e r a t u r e a n d shows t h e need for sure on the oil industry the world’s present produca r a t i o n a l a t t i t u d e toward s u l f u r in gasoline. It to limit the sulfur contion of crude oil contains 0.4 reveals the f a c t that over e i g h t billion gallons of gasotent of gasoline to 0.1 per cent per cent s u l f u r o r m o r e , line, m o r e than two-thirds of t h e country’s t o t a l conlest automobile motors be inMiller6 states: s u m p t i o n , are used in w a r m weather, when according jured by corrosion. Recent The question of desulfurizing t o all published evidence t h e r e is n o danger of corrosion. research under the aegis of gasoline is confronting the reI t seems a b s u r d to refine this h u g e a m o u n t of gasoline the General Motors Corporafinery engineer today in a more to conform t o an a r b i t r a r y s u l f u r specification which in acute form than ever before. tion3 and the Standard Oil The increased producing possimost s t a t e s t h r o u g h o u t the greater part of t h e year Company of Indiana4has inb i l i t i e s of the Permian Salt appears t o be worthless a n d t o serve only to necessitate dicated, however, that corroBasin area of West Texas, with wasteful refining practices.? We urge consideration sion takes place only when a present daily production of of the of t h e only reasonable procedure-regulation the atmosphere is a t a tem350,000 bbls., and an estimated potential production of 750,000 s u l f u r tolerance i n gasoline by t h e climatic conditions perature of approximately 32 O bbls., constitute a serious probwhere it is t o be used. At least i n w a r m weather, F. or lower. Corrosion, it is lem for the future. we see no reason why the s u l f u r l i m i t a t i o n m a y n o t s t a t e d , does not occur in well be revoked entirely, as no o n e has reported a n y Howee writes: warm weather. Why, thereevidence of corrosion u n d e r mild t e m p e r a t u r e condifore, refine gasoline to an exPresent refining methods for tions. treme limit of sulfur the year treating high-sulfur crudes reround? sult in Dolvmerization. the deThe sharr, increase in the amount of high-sulfur crude oils struction of certain sulfur compounds, Hnd unfortunately, the being has made it seriousj,roblem than ever partial destruction of the unsaturated compounds which form such a large and essential part of “cracked” gasoline and which to meet the 0.1 per cent sulfur in gasoline requirement. This regard as valuable in the anti-knock type of gasoline.*** will be seen from the estimate made by a Prominent petroleum A gasoline Droducer estimates that if a higher percentage of sul-

T

Received June 16. 1928. Egloff and Morrell, Oil Gas J . , 26, 1.50 (1927) a Mougey, IND. END.CHEM.,20, 18 (1928). 4 Diggs, I b i d . , 90, 16 (1928). 1

fuicould be tolerated in the gasoline of the-country refining losses

2

A v c r a y e gasoline conaumplion p e t

5

8

Oil Gas J . , 26, 121 (1928). IND. ENG.CHEM.,19, 189 (1927)

motor vehicle bu atatco in 1927. aa indicated by i a z returns. Figurea a r e gallons

840

I N D U S T R I A L A N D ENGINEERING CHEMISTRY

Vol. 20, No. 8

Table I-Sales of Gasoline, by Months, in 1927 (In thousands of gallons) STATE TAN. FEE. MARCH APRIL MAY Turn TULY Auo. SEPT. OCT. NOV. TOTAL DEC. Alabama 9,276 9,437 10,915 11,453 12,462 147,225 14,091 13,358 12,501 Arizona 2,626 2,595 2,812 3,017 3,061 41,237 4,283 4,321 4,521 Arkansas 7,156 6,493 6,280 8,900 6,208 99,394 7,969 9,170 8,216 California 218,816 270,333 254,778 1,017,681 Colorado 8,157 7,482 8,590 7,813 14,814 10,844 10,352 129,184 8,972 Connecticut 8,121 7,717 10,063 12,863 14,474 15,091 13,917 12,406 159,953 Delaware 1,179 1,589 1.716 1,476 1,807 2,089 2,167 24,273 1,946 District of Columbia 3,879 3,619 4,422 4,841 5,313 5,223 4,839 57,804 4,750 Florida 25,119 22,597 23,844 21,807 20,741 247,248 17,285 19,382 20,944 Georgia 16,370 12,486 12,412 15,609 16,771 192,262 18,075 17,250 17,090 2,231 Idaho 1,430 3,671 1,944 3,084 43,343 4,559 3,565 2,668 Illinoisa 50,100 43,700 43,800 55,400 60,600 705,192 66,325 59,564 55,184 Indiana 22,576 19,394 19,457 25,256 27,793 349,995 34,703 33,738 30,435 Iowa 16,847 15,249 18,381 42,049 25,381 309,098 27,515 25,432 19,650 Kansas 16,619 16,075 19,687 14,419 23,818 270,615 23,096 21,512 19,182 Kentucky 6,103 6,046 8,585 10,299 7,484 118,273 11,745 10,384 9,354 Louisiana 11,672 10,640 11,785 12,440 12,346 151,605 14,117 13,414 13,194 Maine 1,947 1,617 1,642 4,463 6,933 74,738 8,163 6,282 3,842 Maryland 7,831 7,406 11,274 9,408 11,111 131,500 12,008 10,916 9,952 Massachusettsb 15,630 14,850 19,300 24,700 27,800 307,180 28,950 26,700 23,850 Michigan 34,180 32,243 40,703 47,404 51,836 593,372 63616 36216 31,340 Minnesota 12,148 14,627 19,728 22,077 25,937 289,686 27:515 24:888 14,515 Mississippi 8,241 8,336 7,499 8,923 9,094 117,756 11,203 11,028 10,385 Missouri 19,783 18,764 23,091 31,058 28,163 322,219 27,982 27,767 24,717 Montana 5,300 11,716 18,079 53,248 Nebraska 10,869 10,386 11,302 11,741 14,867 15,457 14,478 12,610 169,677 Nevada 618 659 925 907 1,047 1,122 1,018 12,720 930 New Hampshire 1,449 1,319 1,757 3,587 3,788 4,686 3,766 45,729 3,253 23,460 22,500 28,066 27,850 31,600 Xew Jerseyc 33,837 31,752 30,165 371,169 New Mexico 2,013 1,985 1,903 2,348 2,543 2,980 2,847 30,654 2,551 55,800 50,900 67,909 77,900 84,700 New Yorkd 89,500 82,250 79,900 968,959 16,291 16,441 14,928 13,715 16,552 North Carolina 219,583 21,834 21,053 20,503 2,276 4,254 4,296 North Dakota 6,324 88,641 7,069 11,119 5,941 2,290 41,631 43,041 54,046 64,460 75,841 Ohio 770,801 71,274 62,721 58,036 16,455 15,762 18,934 19,147 21,645 Oklahoma 251,462 22,032 21,228 21,525 6,564 6,973 Oregon 8,974 10,469 11,337 130,885 11,944 10,291 9,478 Pennsylvania 131,456 691,557 182,147 173,629 3,576 3,182 Rhode Island 4,109 4,976 61,436 6,406 5,367 4,984 4,735 South Carolina 6,551 6,379 7,421 101,772 8,415 8,173 9,055 9,627 9,665 5,135 4,572 South Dakota 6,791 88,031 6,915 5,158 8448 7403 3,841 9,156 8,235 10,299 11,667 13,009 Tennessee 148,952 14:041 12:932 12,456 41,961 38,413 45,803 45,092 49,185 Texas 591,447 53,660 54,131 50,985 2,630 2,025 2,360 Utah 3,511 41,774 2,884 3,594 4,388 3,227 1,225 Vermont 888 877 2,252 3,622 33,167 1,946 3,643 2,378 9,252 9,504 11,783 13,184 14,380 Virginia 166,782 16,703 14,405 13,945 11,205 12,310 14,433 16,870 17,803 Washington 203,421 17,920 16,252 14,983 West Virginia 4,375 4,526 6,186 102,655 7,552 8 795 10641 9,076 7,244 Wisconsin 12,887 13,820 18,046 24,351 29:200 313,586 30:790 26,000 19,136 1,394 1,282 Wyoming 26,218 1,433 2,113 1,754 1,611 2,437 2,070 11,58j8152 Total reported consumptio,n 12,512,976 Total gasoline produced in U. S. for domestic: use 92.6% Proportion covered in this surveyd a Total consumption estimated by X a t l . Petvoleum News, 20, 29 (March 30, 1928). Sales during each of the first 7 months estimated from this total and Indiana statistics. b Total consumption estimated by r a i l . Petroleum News, 20, 29 (May 30, 1928). Sales during each of the first 7 months estimated from this total and Connecticut and Pennsylvania statistics. E Total consumption estimated from motor vehicle registration and average gasoline consumption per car in the United States (500 gallons per year) monthly sales estimated from this total and from Pennsylvania, New Jersey, and Connecticut statistics. d U. S. Bureau of Mines, from Nail. Petroleum News, 20, 29 (May 30, 1928). T h e gasoline unaccounted for includes that exempt from tax in the various states and t h a t used in the service of the Federal Government.

could be reduced with a net annual gain of gasoline worth fifty million dollars. S u m m a r y of Conclusions regarding Corrosion

The work that has been done on corrosion of motors, particularly by Diggs and Mougey, has established the following generalizations: 1-Corrosion is exclusively a cold-weather problem. This trouble occurred only in cold weather-never in summer. The first tests, made in November, were performed during a period of relatively warm weather, on a blend***running about 0.151 per cent sulfur. Exceedingly little corrosion occurred.*** Marked corrosion occurred only when the temperature was in the neighborhood of 32' F. or 10wer.~ There were no complaints in summer regardless of the sulfur content of the fueL3 There is no complaint during the summer, thanks to the beneficial effect of the higher temperature a t which the engine runs and, what is more important, a t which i t s t a r t s 6

2-Corrosion occurs only when water collects in the crankcase. There is almost unanimous agreement that water in the crankcase is responsible [for corrosion], Although water alone will cause corrosion, the action may be accelerated by the formation of weak sulfurous or sulfuric acid. The chemical activity from this [sulfur] and other contaminants would be negligible if effective control of the formation of water were p o s ~ i b l e . ~ It is well known and easily demonstrated that sulfur dioxide and sulfur trioxide will not cause corrosion in the absence of 7 Mougey, footnote 3, quoting Thorne, J. SOC.Az61omotiue En&, 18, a5 (1926).

moisture when the engine is operating a t an optimum temperature. Carbon disulfide has been used experimentally in an automobile engine where there was no water vapor present, and no corrosion took place.6

A number of refiners have made tests with motor fuel containing as high as 0.6 per cent sulfur without any evidence of corrosive action in the motors. 3-Corrosion does not take place, even in winter, in engines in constant use. On engines operated in continuous service in moderately heavy duty, or in long runs such as road tests, no water was found in the crankcases, and the w e k on the cylinders and other rubbing surfaces was much below normal. These conditions will exist with fuels containing up to 0.25 per cent of sulfur, and usually on such cars it will be impossible to show any relation between the percentage of sulfur in the fuel and the wear on the rubbing p a r k 3 When a car was given a 40,000-mile road test [in winter, with no special devices to prevent condensation of water] using a motor fuel containing about 0.25 per cent sulfur, no evidences of corrosion of the rubbing surfaces were found, and the actual wear of the engine parts was very much less than for similar cars in normal private use.3

4--When condensation is not prevented, corrosion occurs even with fuels containing less than 0.1 per cent of sulfur. In a fleet of two hundred taxicabs of the same make, in the winter of 1925-6, althoukh the fuel averaged less than 0.10 per cent sulfur, and probably the maximum sulfur in any lot of the fuel was 0.12 per cent, very severe corrosion troubles were experienced on practically the entire lot of these cars.

INDUSTRIAL AND ENGINEERING CHEMISTRY

August, 1928

M i n i m u m T e m p e r a t u r e s by M o n t h s (Degrees Fahrenheit) M A Y JUXE M A R C H APRIL JULI AL-GUS? SEPT. 62 56 48 69 71 71 66 45 60 51 40 69 67 60 67 59 50 41 70 69 62 47 45 55 51 59 59 56 41 50 23 32 56 55 46 49 57 64 62 28 38 55 61 53 43 68 33 59 66 54 43 68 63 66 59 34 64 69 75 73 73 75 50 62 68 54 4: 71 67 71 48 51 33 38 58 48 56 54 43 69 63 33 58 66 54 43 66 62 58 6.5 33 60 40 51 65 62 2: 55 61 43 53 67 66 33 58 46 65 56 68 60 61 36 59 66 72 74 70 74 50 42 32 50 20 59 49 55 64 55 45 69 61 67 35 48 38 63 56 62 29 56 44 34 53 21 59 51 57 44 34 60 54 58 19 50 60 70 68 46 53 64 70 45 64 55 59 68 66 34 41 48 33 23 48 55 51 49 59 64 62 38 42 25 41 49 34 46 56 54 29 44 52 33 23 48 59 55 61 52 42 67 33 60 66 47 39 56 61 33 52 59 47 62 36 56 24 53 60 58 49 66 42 62 70 69 41 51 13 26 43 56 53 41 52 61 65 57 64 30 66 58 40 49 69 63 69 44 39 47 48 52 3fj 51 40 60 51 65 30 57 63 47 38 61 63 62 2!) 57 61 45 52 71 68 69 70 46 35 61 56 21 50 59 59 66 4.! 50 62 70 69 64 56 71 50 67 73 73 42 51 36 48 59 20 58 44 52 32 48 59 55 19 55 45 67 63 61 67 38 45 40 55 50 49 315 53 62 51 66 72 70 4.? 73 47 36 62 57 23 53 60 37 50 48 45 29 1‘) 40

841

T a b l e 11-Mean STATE Alabama Arizona Arkansas California Colorado Connecticut Delaware District of Columbia Florida Georgia Idaho Illinois Indiana OW -T _ . _l

a

JAN.

FEE.

39 33 30

400 36 32

40

42

14 20 25 26

10 22 24 27

39 22 21 22 11 15 26

26 22 22 13 22 27

53

55 40

Kansas Kentucky 43 46 Louisiana 10 10 Maine Maryland 27 28 Massachusetts 23 22 Michigan 13 12 1 4 Minngsota Mississippi 37 38 Missouri 21 23 Montana 9 11 Nebraska 12 14 21 24 Nevada 12 11 New Hampshire New Jersey 26 25 New Mexico 22 25 New York 16 15 North Carolina 35 35 6 4 North Dakota Ohio 21 22 Oklahoma 28 28 Oregon 31 33 22 22 Pennsylvania Rhode Island 23 21 South Carolina 37 37 6 8 South Dakota Tennessee 32 34 40 42 Texas Utah 19 23 Vermont 8 8 Virginia 30 30 Washington 31 33 33 32 West Virginia Wisconsin 9 11 9 11 Wyoming Temperatures of 40‘ F. or over are in italics.

The crankcase oils were badly contaminated with water and iron oxide.3

5-Corrosion may be prevented by equipping c:trs with devices for the regulation of jaAiet-water temperature and ventilation of the crankcase, these improvements eliminating the condensation of water. On the lot of taxicabs already mentionedin the spring of 1926***thermostats and ventilators [were] installed. This change absolutely prevented corrosion troubles in the next winter.***This make of car in private service had been subject to corrosion troubles, but after the adoption of the ventilator and thermostat not a single complaint on corrosion was received from the entire year’s production, regardless of the fact that many of these cars were operated on fuel s ith a sulfur content of approximately 0.25 per cent.3

With another group of cars with no preventatives of condensation of waterwith fuels with about 0 2 5 per cent of sulfur the complaints were very numerous and very bad, especially in the case of cars in intermittent service. In the 1925-6 series, with thermostats, ventilators, and oil filters, water in the crankcase was eliminated, and no corrosion complaints were received, regardless of the sulfur content of the fuel.3

Gasoline Consumption Data

Since these authorities show that corrosion occurs only under winter conditions and then only in cars in intermittent service not equipped to prevent condensation of water in the crankcase-the amount of sulfur which may be allowed in gasoline becomes primarily a question of climatic conditions. To give some idea of the magnitude of gasoline consumption in “non-corroding” weather three tables are presented shomring gasoline sales throughout the country in relation to the lowest temperatures usually encountered.

OCT.

Nov.

DEC.

55 48 50 51 35 45 48 47 67 56

45 39 40 45

39 32 32

41

24 34 38 37

16 25 28 28

39

47 47 43 45 48 60 40 49 47 41 38

53 47 33 38 36 39

50 41 43 54 32

46 50 42 46 48 55 37

51 59 37 38

50 43 56 42 30

59 46

31 35 38 30 33 37

54 40 23 25 25 18 23 29

50

44

29 39 37 28 23

17 30 28 19 9 37 25 15 17 22 17 29 22 22 36 2 26 30 33 26 33 37 13 34

43 35 22 26 28 28 39 30 32

43 17 35

40

37 36 38

44 23

41 49 30 27 39 37

43 26 20

41

20 14 32 33 34 16 11

Table I shows the gasoline sales in 1927 by months, in the various states. The figures are from reports of the state taxing authorities.s Since in some states there is no gasoline taxation or inspection, exact figures on their use of gasoline are not obtainable. For these states estimates have been made by methods indicated in the footnotes. I n connection with this and the succeeding tables, the map showing the gasoline consumption per car in the different states in 1927 is of considerable interest. Temperatures in Various States

In Table I1 are shown, by months, the mean minimum temperatures in all states. The data are taken from publications of the U. S. Weather Bureau covering observations made a t their two hundred and eight regular stations over periods of from five to more than fifty years, the figures usually including a t least twenty years. The stations, so disposed as to give a representative picture of the climate of each state, are located in the following cities: STATE Alabama Arkansas California Colorado Connecticut Delaware District of Columbia Florida Georgia Idaho Illinois Indiana

CITIES Anniston Birmingham Mobile Montgomery Bentonviile, Fort Smith Little ’Rock Eureka, Fresno, Indepbndence, Los Angeles, Red Bluff, Sacramento, San Diego, San Luis Obispo, San Francisco Denver, Grand Junction, Pueblo, Lead>ille New Haven Wilminaton Washington Jacksonville, Key West, Miami, Pensacola, Tampa, Titusville Atlanta, Augusta, Macon, Savannah, Thomasville Boise, Lewistown, Pocatello Cairo, Chica 0,Peoria, Springfield Evansville, s o r t Wayne, Indianapolis, Terre Haute

8 Am. Petroleum Inst., 9, No. 20 (March 21, 1928); N o & Petroleum News, 20, 29 (May 30, 1928).

INDUSTRIAL AND ENGINEERING C H E - I S T R Y

842 STATE Iowa Kansas Kentucky Louisiana Maine Maryland Massachusetts Michigan Minnesota Mississippi Missouri Montana Nebraska Nevada New Hampshire New Jersey New Mexico New York North Carolina North Dakota Ohio Oklahoma Pennsylvania Rhode Island South Carolina South Dakota Tennessee Texas

Utah Vermont Virginia Washington West Virginia Wisconsin Wyoming

CITIES Charles City, Davenport, Des Moines, Dubuque, Keokuk, Sioux City Concordia, Dodge City, Iola, Topeka, Wichita Louisville, Lexington New Orleans, Shreveport Eastport, Greenville, Portland Raltimore -... Boston, Nantucket Escanaba, Houghton, Marquette, Sault St. Marie Duluth, Minneapolis, Moorhead, St. Paul Corinth, Jackson, Meridian, Vicksburg Columbia, Hannibal, Kansas City, Springfield, St. Joseph, St. Louis Billings, Havre, Helena, Kalispell, Miles City, Missoula Lincoln-, North Platte, Omaha, Valentine Reno, Tonopah, Winnemucca Concord Atlantic City, Cape May City, Sandy Hook, Trenton Roswell, Santa F e Albany, Binghamton, Buffalo, Canton, Ithaca, New York, Oswego, Rochester, Syracuse Asheville, Charlotte, Hatteras, Manteo, Raleigh, Wilmington Bismarck, Devils Lake, Grand Forks, Williston Cincinnati Cleveland, Columbus, Dayton, Sandusky, Toledo Muskegee Erie, Harrisburg, Philadelphia, Pittsburgh. Reading, Scranton Block Island, Providence Charleston ~ ~, Columhia. _.._ ~ ~ ~ ~. ~Greenville .. ~ ~ . ,. Huron, Pierre, Rapid City, Yankton Chattanooga, Knoxville, Memphis Abilene Amarillo Brownsville Corpus Christi Dalds, Del Rio,’El Paso, Fort’Worth. Galveston: Houston, Palestine, Port Arthur. San Antonio. Taylor ’ Modena, Salt Lake City Burlington, Northfield Cape Henry, Lynchburg, Norfolk, Wytheville Blaine North Head Port Angeles Seattle Spokane, TatAosh Island, $acorns, Walla’ Walla, Yakima Elkins, Parkersburg Green Bay, L a Crosse, Madison, Milwaukee Cheyenne, Lander, Sheridan, Yellowstone Park ~

~~~

~

~___

~~~~~~

~~~~

T o obtain the data in Table 11, a t each station, the lowest temperature for each day was recorded. The observations made during a month by each station were averaged, giving the monthly mean minimum temperature. An average of these for the number of years the station had been in operation was then made, and the means for all the stations in each state were again averaged to give the state’s monthly mean minimum temperatures, which are the figures used in the table. As they show the lowest temperatures normally found, these mean minimum temperatures are the best available index to weather conditions as they affect the sulfur problem. A mean minimum temperature of 40” F. has been selected as a suitable dividing line between weather requiring lowsulfur gasoline and that in which a sulfur specification should be unnecessary. I n view of the statement of Diggs, that “marked corrosion occurred only when the temperature was in the neighborhood of 32” F. or lower,” the choice of 40’ F. gives an ample margin of safety for irregularities in temperature. I n the table figures of 40” F. and above are therefore italicized to denote periods in which the published data indicate that corrosion will not occur. The mean temperature in any state is, of course, much higher than the mean of the minima here given. Sales of Gasoline i n Warm and Cold Weather

To show the relative proportions of warm- and cold-weather gasoline, the sales of gasoline in 1927 in months whose mean minimum temperatures are 40” F. or above, and the sales in months below this temperature, are given in Table 111. It will be noted that approximately 71 per cent of the gasoline consumption reported was in weather warm enough to prevent corrosion. A number of states should experience no corrosion on high-sulfur gasoline a t any time of the year, notably California and Texas, both of which consume large quantities of gasoline and produce high-sulfur crude oil in large volume. To be sure, there are mountainous areas t o which the climatic generalizations will not apply. But in view of the published data on corrosion, the table would

Vol. 20, No. 8

indicate that for a large part of the gasoline marketed a sulfur specification is superfluous. At first thought it may not seem feasible to fit gasoline to the weather. But the differentiation of summer and winter gasoline is now established practice. In volatility the Bureau of Mines’ semiannual surveys show definite seasonable variation. There should be no insurmountable difficulty in distributing gasoline without regard to sulfur content in the summer months, and reducing the sulfur, in addition to raising the volatility when cold-weather conditions prevail. of G a s o l i n e in W a r m a n d Cold W e a t h e r MEAN MIN. MEAN TEMP. SALES PART MIN. SALES 40’F. IN OF TEMP. IN OR WARM TOTAL BELOW COLD ABOVE MONTHS SALES 4 0 ’ F . MONTHS STATE I000 Per 1000 Months gallons cent Monlhs gallons Alabama 10 129,512 87.3 2 18,713 27,174 65.9 4 Arizona 8 14,063 77,529 78.0 Arkansas 9 3 21,865 100.0 California 12 1,017,681 0 0 46.9 60,750 Colorado 5 7 68,434 58.1 6 67,087 92,866 Connecticut 6 5 15,676 Delaware 7 64.6 8,597 5 36,295 63.5 District of Columbia 7 21,509 100 Florida 12 247,248 0 0 85.2 1 Georgia 16,370 ‘1 175,892 52.7 Idaho 20,481 22,862 7 64.2 452,842 5 Illinoisa 2 52,360 64.1 224,395 5 Indiana 125,600 7 213,539 69.1 5 7 Iowa 95,559 177,540 66.3 5 7 Kansas 93,075 78,902 66.6 5 7 Kentucky 39,371 100 151,605 12 0 0 Louisiana 72 1 Maine 53,945 6 6 20,793 84,121 63.0 5 7 47,379 Mary 1and 183,050 59.5 6 124,130 6 Massachusettsb 370,286 62.4 6 6 Mi chi gan 223,086 53.2 154,188 5 7 135,498 Minnesota 76.8 91,536 9 3 26,220 Mississippi 64.6 5 208,097 7 114,122 Missouri 51.1 7 27,206 5 26,042 Montanac 82,834 48.8 7 5 86,843 Nebraska 51.4 7 6,179 6,541 5 Nevada 5 56.8 7 19,727 25,995 New Hampshire 7 63.5 5 135,943 235,226 New Jerseyd 55.5 6 6 13,647 New Mexico 17,007 57.2 6 6 414,659 554,300 New Yorkb 75.8 3 166.348 9 53,235 North Carolina 7 52,141 58.8 5 36,500 North Dakota 5 511,326 66.5 7 Ohio 259,475 9 3 197,720 72.8 53,742 Oklahoma 6 52,749 78,136 59.4 6 Oregon 249,353 442,204 63.9 5 7 Pennsylvania e 6 25,562 35 874 58.1 6 Rhode Island 3 9 21,985 79:787 78.4 South Carolina 5 41,348 46,683 53.0 7 South Dakota 119,005 75.2 9 29,847 3 Tennessee 12 0 0 591,447 100.0 Texas 20,749 49.6 5 7 21,025 Utah 19,958 60.5 5 13,209 7 Vermont 5 101,893 64.7 58,889 Virginia 7 134,238 66.0 69,183 7 5 Washington 16,145 86,510 81.1 9 3 West Virginia 199,346 63.6 114,240 6 6 Wisconsin 12.124 46.2 14,094 4 8 Wyoming T a b l e 111-Sales

;



i n 1927 PART OF

TOTAL SALES Per cent 12.7 34.1 22.0 0.0 53.1 41.9 35.4 36.5 0.0 14.8 47.3 35.8 35.9 30.9 33.7 33.3 0.0 27.9 36.0 40.5 37.6 46.8 23.2 35.4 48.9 51.2 48.6 43.2 36.5 44.5 42.8 24.2 41.2 33.5 27.2 40.6 36.1 41.9 21.6 47.0 24.8 0.0 50.4 39.5 35.3 34.0 18.9 36.5 53.8

~

8.197.229 Av. 7 0 . 8 3,387.923 Av. 2 9 . 2 Totals a See footnote a , Table I. b See footnote b, Table I. c Figures for sales by quarters furnished by state; sales by months estimated from sales in corresponding months in North and South Dakota. d See footnote c, Table I. e Sales by quatters reported by state; sales by months estimated from sales in Xew jersey

Conclusion

This proposed elasticity in the sulfur specification may meet objection on the ground that allowing the marketing of high-sulfur gasoline for even part of the year could lead to abuse of this privilege. But there is even now no bar to the selling of any sort of gasoline, except the integrity of reputable oil companies mho may be relied upon not to play false to the interests of the motoring public. The improvement in gasoline in recent years, particularly the introduction of fuel of high volatility during the winter months, has not been primarily in response to a demand from the public, but is an advance due to competitive conditions and a real desire to market the best product possible. The same forces would control the putting into practice of changes in the sulfur specification. It would be a great improvement over

August, 1928

INDL'STRIAL A N D ENGINEERING CHEMISTRY

existing conditions if the present unnecessarily rigid standard were replaced by regulation of the sulfur tolerance in gasoline in accord with the -known facts regarding corrosion. We are not urging that no attention be Daid to the sulfur content of motor fuel. But the elimination of the sulfur requirement in warm weather is in accord with published work on corrosion, and would be a means of effecting economies in refining and of supplying better antiknock motor

843

fuel. The annual saving to the country which this measure would accomplish is estimated at thirty-five million dollars. Acknowledgment

The writers wish to express their appreciation of the assistance given by B. Parry, of the New York office of the U. S. Weather Bureau, in obtaining the temperature data used in this paper.

Lubrication Round-Table Discussion D

P. BARNARD, 4th-Those engaged in supplying lubri- Parsons and Taylor, and the substitution of a brine bath cants for automobile equipment know that the prob- for the freezing mixture previously employed. lems of warm-weather lubrication are by no means so seriThe results obtained by this method are illustrated by ous as those encountered when cars are operated a t low Figure 2, in which the rates of flow have been plotted against temperatures. Some time ago this laboratory undertook the pressure drop for three typical oils. The logarithmic an experimental program for the purpose of determining the scale was used in this work instead of arithmetical coordiviscosities of typical automotive lubricating oils a t tem- nates, as employed by Parsons and Taylor, to afford constant peratures down to - 17" C. (0°F.) and to study the effects of precision in plotting and to offer a wider range of observations. On such a plot any viscosity and pour test upon fluid which follows the laws the resistance to cranking of v i s c o u s flow, or flow offered by the oil and upon The Round-Table Discussion on Lubrication was d i r e c t 1y proportional to its circulating characterisarranged as a joint meeting of the Petroleum and Inpressure, will give a straight tics. The results indicate dustrial and Engineering Chemistry Divisions at the St. line having a slope of 1. that cranking effort is deLouis Meeting of the American Chemical Society in order However, none of these oils termined almost entirely by to provide for the informal presentation of the progress showed such a slope a t the the viscosity of an oil apart made since the Lubrication Symposium at Tulsa two l o w e r temperatures indifrom its cold test, while ciryears ago. R. E. Wilson was appointed chairman of cated. culation by the lubricating both meetings, but was unable to attend the one at The deviation in the case s y s t e m d e p e n d s largely St. Louis. In his absence D. P. Barnard, 4th, served of the asphalt-base oil was upon the temperature of the as a substitute and opened the discussion by describing not great, but in the other oil being above its pour test. some of the experimental work recently carried out by The first work to be pubtwo it was marked, the flow the Standard Oil Company of Indiana. i n c r e a s i n g more rapidly lished on the subject of oilthan in proportion to the viscositv characteristics a t low temperatures was that p r e s s u r e . The data for of Parsons and Taylor,' who presented the results obtained glycerol have been included for the purpose of demonstratfrom two widely varying types of oil, asphalt- and paraffin- ing that the non-conformity with Poiseuille's law was not in base, when compared in a Saybolt viscometer under varying apparatus characteristic. The same data plotted as apparent heads. Their results are illustrated in Figure 1, in which the viscosity us. pressure-drop are shown in Figure 3. The slopes rate of flow has been plotted against the corresponding pres- of these graphs become equal to 1-8, where 17 is the slope of the lines in Figure 2 . sure drop. It will be noticed that the asphalt-base oil apparently Departure from Laws of Viscous Flow conformed to Poiseuille's law in that the flow varied directly The departure from the ordinary laws of viscous flow, as with the pressure drop. The paraffin-base oil, however, behaved quite differently in that the flow increased more illustrated in Figures 2, 3, and 4, is typical of the different rapidly than the pressure drop, indicating that the oil a t classes of oils examined. It is evidently due to the presence this temperature was a plastic material the apparent vis- of a colloidal structure which becomes more pronounced as cosity of which decreased with increasing shearing stress. the temperature is lowered. I n the case of the paraffin-base This characteristic is most important, as it indicates that two oils the solid phase is without doubt mainly wax crystals, but oils may have the same viscosity at 38.5" C. (100" F.) but some kind of colloidal particles also appear to be present in entirely different properties at - 17" C. (0" F.) The paraffin- the asphalt-base oils. Figure 4 shows the results of several observations at varibase oil, for instance, although having a comparatively high pour test, actually flowed more readily a t high rates of shear ous temperatures for oils of different viscosities belonging than did the asphalt-base oil. I n other words, while the to the three general types. It was observed throughout this paraffin-base oil had the higher apparent viscosity at low work that 8 is rather erratic, although variations were never rates of shear, this condition was reversed at high rates of so great as to cause overlapping of oils of the different classes. shear. Parsons and Taylor did not, however, describe any It is probable that deviation of 7 from unity is in some way correlation between these results and actual engine per- related to the pour-test temperature as specified by the American Society for Testing Materials, but the data availformance. A similar method was followed in the viscosity experiments able a t present do not indicate the existence of any definite a t Whiting, the chief differencein procedure consisting in the relation and evidently the deviation persists well above the use of thermocouples placed a t the entrance to the metering pour point. Furthermore, there seems to be no direct bearcapillary in place of the thermometer and dummy tube of ing of viscosity within the limits examined, upon the value of 7. In general, this work can be summarized in accordance with Table I. 1 IND. ENG CHEM,18,493(1926).